ResearchTasksMultitask Language EvaluationMultitask Language EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedArabic Benchmarks (COPA-ar, HellaSwag-ar, ArabicMMLU)AceGPT-7B-chat69.7COPA-ar7May 29, 2026